The pteridines utilized by higher animals fall mainly into two classes of pterin (2-amino-4-hydroxypteridine) enzyme cofactors, the derivatives of poly-gammaglutamyltetrahydrofolates and tetrahydrobiopterin, although several pteridines diverging widely from the substitution patterns of these pterin series have significant roles in bacteria, plants, and insects. A common factor of several of the tetrahydropterin-utilizing enzymes is the oxidation of the pterin cofactor. For example, in the case of thymidylate synthetase (a major target of antifolate chemotherapy), the 5,10-methylene derivative of tetrahydrofolic acid is converted to 7,8-dihydrofolic acid. Likewise, the action of the three aromatic amino acid hydroxylases (phenylalanine, tyrosine, and tryptophan) yields a quinoid dihydrobiopterin of uncertain structure.
Catalysis by tyrosine hydroxylase is the rate limiting step in the biosynthesis of dopamine and norepinephrine. The activity of this enzyme in the caudate nucleus and substantia nigra is markedly lower in Parkinson's disease as is also the concentration of its cofactor tetrahydrobiopterin. Parkinson's disease is a debilitative disease which usually appears insidiously between 50 and 60 years of age. The disease is progressive, generally beginning with tremor followed by bradykinesia and rigidity. Estimates of the current number of cases of Parkinson's disease in the United States range between 500,000 and one million. Approximately one in 40 individuals will eventually become afflicted, creating about 50,000 new cases per year. A variety of anatomical and biochemical defects in the brains of Parkinson's patients have been noted, the most prominent involving the basal ganglia, with cell loss and depigmentation in the pars compacta of the substantia nigra. A large part of this lesion is due to degeneration of dopaminergic neurons in the nigro-striatal pathway.
A marked deficiency in dopamine concentration (less than 10% of normal, decreasing further with time after onset) is seen in striata of Parkinson's patients. Since its introduction in the mid-1960's, the use of L-dopa has become the major means of treatment. L-Dopa enters the brain via the aromatic amino acid transport system and is subsequently enzymatically decarboxylated to dopamine. Unfortunately, the action of peripheral decarboxylases normally convert about 95% of administered L-dopa rapidly to dopamine, which does not effectively permeate the blood-brain barrier. As a result, patients are frequently given a combination of carbidopa (a decarboxylase inhibitor) and L-dopa (Merck, Sharp and Dohme-Sinemet), thus reducing the total L-dopa required for optimal effect by about 75%.
Although L-dopa is generally considered the best available treatment for Parkinson's disease, a number of problems are encountered in its use. For example, between 15 and 25% of patients are totally unresponsive to any regime of L-dopa therapy. Furthermore, most patients cannon immediately tolerate the optimal dose and must be gradually titrated to an individual level. Even then, most experience nausea, especially during the initial phase of dosage increase. This adjustment period can be shortened to between 2 and 4 weeks by simultaneous use of carbidopa. The most serious common adverse reactions to L-dopa are abnormal involuntary movements and behavioral disturbances. Within the first 2 to 4 months of therapy about half of patients display choreiform or dystonic movements, increasing to about 80% of those on full dosage schedule for over a year. Serious mental side effects (psychotic episodes, depression and dementia) requiring reduction or withdrawal of the drug is seen in about 15% of cases. Although occasionally moderated by combination with carbidopa, many patients experience swings in ability of L-dopa to suppress bradykinesia (the "on-off" phenomenon).
The full benefit of L-dopa therapy generally lasts for only 2 to 4 years, followed by a decline in response. The effective duration or each dose decreases and akinesia paradoxica or "freezing" increases. Within 5 years the increasing side-effects usually begin to outweigh the remaining benefit of continued treatment. Since most Parkinson's patients live for 10 to 20 more year after onset of the disease, a need clearly still exists for a more effective and long lasting approach. A number of agents have been tested, either as modifiers of dopa therapy, or as dopaminergic agonists to replace L-dopa. At best, these approaches result in a trade-off between improvement of one side effect and the worsening of another. Accordingly, new methods of treating Parkinson's disease are needed.
Along these lines, several attempts have been made to increase cofactor concentration in the brain of patients having insufficient tyrosine hydroxylase activity and thus to increase the race of tyrosine hydroxylation and dopa synthesis. Direct intraventricular infusion of tetrahydrobiopterin (BH.sub.4) into rat brains caused a concommitant increase in catechols in the striatum with increase in brain concentration of tetrahydrobiopterin (Ketter et al., Nature, 249 476-478 (1974)). The effectiveness of BH.sub.4 treatment has been suggested to be possibly even greater in Parkinson's patients, who have a two-fold lower than normal level of BH.sub.4 in the cerebrospinal fluid (CSF) and four-fold lower in the brain (Lovenberg et al., Science, 204, 624-626 (1979)). However, peripherally administered tetrahydrobiopterin (BH.sub.4) does not readily enter the brain. In the first experiments which demonstrated the poor ability of BH.sub.4 to cross the blood brain barrier, 125 mg BH.sub.4 /kg body weight was injected intravenously into rays. BH.sub.4 in the striatum was increased slightly and transiently, with a maximum increase of 30% in 15 mins, returning to normal in less than 2 hours. This is equivalent no 0.2% of the concentration administered assuming uniform body distribution. (Kettler et al., supra.)
After i.p. injection of .sup.14 C-BH.sub.4 into rats, the isotope was detectable in all tissues analyzed (brain, kidney, liver, plasma, urine). Radioactivity peaked in the brain 1 hour after injection, at which time it was less than 1% of that found in plasma. BH.sub.4 was lost from the plasma somewhat more rapidly than from the brain, but by 6 hours most of the radioactivity appeared in the urine as biopterin and its metabolites. Of the 64 .mu.g of radioactive pterin injected per rat, about 10 ng (0.016%) reached the brain (Gal et al., Neurochem. Res., 1 511-523 (1976)).
A comparison of oral and i.v. administration of BH.sub.4 and other pterins has been made in humans, 1-5 years of age, with a genetic disorder in BH.sub.4 biosynthesis. In addition to the neurological symptoms due to the inability of tyrosine and tryptophan hydroxylases to function in the absence of BH.sub.4, these patients also have plasma phenylalanine levels 10 to 20 times normal, due to the requirement of liver phenylalanine hydroxylase for BH.sub.4. It was found that doses of 2.5 mg BH.sub.4 /kg body weight administered by either route decreased the plasma phenylalanine levels to normal within 3-4 hours and maintained these low levels for 1-2 days. Dihydrobiopterin and sepiapterin, the presumed immediate precursors of BH.sub.4 were as effective as BH.sub.4 in lowering plasma phenylalanine when administered orally in doses of 2.5 mg/kg and 0.6 mg-1.25 mg/kg, respectively, and maintained normal levels of phenylalanine for 24 hours (Schaub et al., Arch. Dis. Childhood, 53, 674-683 (1978); Niederwieser et al., Lancet, 131-133 (1979); Curtius et al., Clin. Chim. Acta, 93, 251-262 (1979)). Therefore, oral administration of biopterin analogues appears to be equally as effective as i.v. administration.
The ability of BH.sub.4 to enter the brain has been studied in BH.sub.4 deficient patients. After 2 mg BH.sub.4 (i.v.)/kg/day for 3 days in a 1-2 yr old, 15-kg body weight, BH.sub.4 -synthesis-deficient child, no increase in CSF BH.sub.4 was detected 1 day after the last injection. Analysis of CSF at this time for dopamine and serotonin metabolites showed a slight increase in homovanillic acid from 29 to 44 ng/ml (normal=130 ng/ml) and in hydroxyindoles from 92 to 147 ng/ml, indicating than a trace of BH.sub.4 had penetrated the brain (Danks et al., Pediat. Res., 13, 1150-1155 (1979)).
At doses &lt;2.5 mg BH.sub.4 /kg body weight there is insufficient penetration of BH.sub.4 into the brain to have any effect on the neurological symptoms (ptosis, ataxia) displayed by BH.sub.4 deficient patients. However, at higher doses, up to 22 mg BH.sub.4 (p.o.)/kg, alleviation of the symptoms has been demonstrated in two patients. With one of these patients, sepiapterin was shown to have an effect at 2.75 mg (p.o.)/kg (Niederwieser et al., Eur. J. Pediatr., 138, 110-112 (1982)). Studies with the other patient demonstrated an increase in CSF BH.sub.4 from 2 ng to 44 ng/ml 2.5 hours after oral treatment at a dose of 20 ng/BH.sub.4 /kg/day (administered in two equal doses at 12 hr intervals). This is 0.22% of that which would be expected assuming uniform body distribution and retention and is close to that noted above in rat brain 1.5 hrs after i.p. injection of 30 mg/kg (Kaufman et al., Pediatrics, 70, 376-380 (1982 )).
Currently 6-methyltetrahydropterin (6-MePH.sub.4) is the only other tetrahydropterin that has been studied by direct analysis for its ability to enter the brain. In rats, administration of 6-MePH.sub.4 to rats, 0.11 .mu.mole (i.p.)/gm body weight, demonstrated that this pterin entered the brain up to 10 times more efficiently than an equal dose of BH.sub.4. Levels of 2 nmoles/g brain were reached (i.e., 2% of that expected assuming equal body distribution) at 30 minutes and were maintained until 2 hours after injection. The half life for retention by the brain was 3 hrs. The blood level was 40 .mu.M at 30 minutes and had dropped to 10 .mu.M by 2 hrs, the half life for retention by the plasma being 0.7 hours. In one experiment, enzymatic analysis demonstrated that over 85% of striatal 6-MePH.sub.4 remained in the fully reduced tetrahydro-form 2 hrs after injection (Kapatos and Kaufman, Science, 212, 955-956 (1981)). However, similar experiments by experimenters who claim the use of more reliable assay techniques indicate that only 30% of the 6-MePH.sub.4 in the brain was in the reduced form (Curtius et al., in Pteridines and Folic Acid Derivatives, Blair, ed., Walter de Gruyter, Berlin, 1982).
The effectiveness of 6-MePH.sub.4 has been tested in one human patient with inherited deficiency of BH.sub.4 biosynthesis. Three hours after an intravenous injection of 20 mg 6-MePH.sub.4 /kg body weight, the CSF level was 0.45 .mu.g/ml (i.e., 2.2% of that expected on the basis of equal body distribution), dropping to 0.06 .mu.g/ml at 9.5 hours. Two hours after a dose of 8 mg/kg, CSF homovanillic acid increased from 9 to 22 ng/ml (normal-132 ng/ml for 2-4 yrs of age), and 5-hydroxyindoleacetic acid increased from 6 to 18 ng/ml (normal=30 ng/ml for 2-4 yrs of age). Similar improvements in clinical symptoms were observed as with BH.sub.4 treatment. (Kaufman et al., (1982), supra).
Another problem associated with the use of tetrahydrobiopterin as its instability. As shown in the following scheme, during hydroxylation BH.sub.4 is oxidized to quinoid dihydrobiopterin (BH.sub.2) which is then reduced back to BH.sub.4 in the presence of NADH and dihydropteridine reductase. The nonenzymatic oxidation of tetrahydropterins by molecular oxygen, a generally rapid process in neutral or alkaline aqueous media, also initially generates the quinoid dihydroform. The quinoid BH.sub.2 form is unstable and in the absence of a reducing system rearranges to 7,8-dihydrobiopterin with a half-life of only a few minutes under physiological conditions. This latter tautomer is not a subtrate for dihydropteridine reductase. ##STR2##
As will become clearer when the present invention is more fully disclosed, the present invention relates to 6,6-disubstituted tetrahydropteridines and related compounds in the treatment of Parkinson's disease and other disorders that would benefit from activation of aromatic aminoacid hydroxylases. Although some 6,6-disubstituted tetrahydropteridines have previously been synthesized for chemical studies, their use and their advantages for such treatments do not appear to have been previously recognized.
The first synthesis of 6,6-disubstituted tetrahydropteridines utilized nucleophilic addition of cyanide across the 7,8 double bond of 6-methyl-7,8-dihydropterin, yielding 6-cyano-6-methyltetrahydropterin (Viscontini et al., Helv. Chim. Acta, 54, 811-818 (1971)). After a number of intermediate steps, the nitrile was reduced and the final compound, 6-aminomethyl-6-methyl-5,6,7,8-tetrahydropterin, was obtained. This molecule proved to be less stable than expected, since upon oxidation the resulting quinoid dihydropterin was able to reform the starting 6-methyl-7,8-dihydropterin by loss of the amino methyl group as ammonia and formaldehyde.
Recently the above approach has been extended to the synthesis of 6,6-dimethyltetrahydropterin by reactin of methyllithium with trimethylsilylated 6-methyl-7,8-dihydropterin followed by desilylation (Armarego and Waring, Aust. J. Chem., 34, 1921-1933 (1981)). A yield of 28% was reported, but evaluation of the ultraviolet absorbance characteristics indicates that it was only 82% pure. Although this method is relatively straightforward, two potential disadvantages are envisioned. First, a chromatography step is required in order to liberate the relatively low yield of desired material from by-products, a process that would inhibit scale up. More importantly, it is doubtful that this procedure has a wide scope, for although a number of 7,8-dihydropterins monosubstituted at position 6 are potentially available by current procedures, it seems likely that yield will further suffer with increasing hindrance of C(6) by groups larger than methyl.
Although many procedures are known for the synthesis of pteridines, the only known prior art relevant to the overall synthetic method of the present invention can be found in a publication of Lazarus et al., Biochemistry, 20, 6834-6841 (1981). The purpose of the relevant part of this work was to determine whether 2,5-diamino-6-(meso-1-methyl-2-aminopropylamino)-4-pyrimidinone could be oxidatively cyclized to fully oxidized 6,7-dimethylpterin. Nowhere in this work is it stated or implied that the procedure is a general method of pteridine synthesis. Further, the potential application to the synthesis of 6,6-substituted tetrahydropteridines is nowhere discussed, since the main thrust of the article is toward preparing a fully oxidized pterin, which would not be possible with disubstitution at a single ring position. Nor is the use of such compounds for the treatment of Parkinson's disease or other diseases involving pterin cofactors discussed in this or any of the other references which disclose 6,6-disubstituted pterins.
Accordingly, prior to the present invention, there remained a great need for an improved method of regulating enzymes having tetrahydropterin cofactors.